#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""Dump the CLASSIFICATION fields the client stored for every card it holds, so
the subtype->cardtype map and the itemState runtime values are read from the
running game instead of inferred.

READ-ONLY. /proc/PID/mem is opened 'rb'; there is no write path in this file.

WHY THIS EXISTS
---------------
Two things this project has repeatedly had to treat as INFERRED:

  1. `FUN_1800d8330`'s cardsubtypeid -> cardtype map. It is read out of Ghidra
     (0..3->1 players, 4->2 manager, 5->3 headcoach, 6->10 gkcoach, 7->5 physio,
     8->4 fitnesscoach, 9..b->7), and the kit selector gate `FUN_1801c3480`
     branches on cardtype == 7. Serving a subtype whose cardtype we guessed
     wrong fails SILENTLY, because cardtype 9 has no arm in the merge.
  2. The itemState enum. The table at 0x180229d20 gives the tokens; the RUNTIME
     values the strings deserialize to (notably activeHomeKit/activeAwayKit ->
     101/102) have been carried as inferred.

Both are directly observable: the parser writes cardsubtypeid to rec+0x50, the
derived cardtype to rec+0x4c, and the decoded itemState to rec+0x5c. Reading
those back for every record turns the pair into measurements.

    rec+0x18  resourceId
    rec+0x4c  cardtype   (derived by FUN_1800d8330 from cardsubtypeid)
    rec+0x50  cardsubtypeid (as sent)
    rec+0x5c  itemState  (decoded enum value)

Usage:  python3 record_vocab_probe.py
"""
import collections
import sys

import watch_club_model as W
import card_identity_probe as P

F_RESOURCE = 0x18
F_CARDTYPE = 0x4C
F_SUBTYPE = 0x50
F_ITEMSTATE = 0x5C
REC_SIZE = 0x158

# What the Ghidra read of FUN_1800d8330 predicts, so a disagreement is loud.
EXPECTED_CARDTYPE = {0: 1, 1: 1, 2: 1, 3: 1, 4: 2, 5: 3, 6: 10, 7: 5, 8: 4,
                     9: 7, 10: 7, 11: 7}


def main():
    pid = W.find_pid()
    if pid is None:
        print("FIFA17.exe is not running.")
        return 1
    base = W.dll_base(pid)
    if base is None:
        print("pid %d is up but %s is not mapped yet." % (pid, W.DLL))
        return 1
    mem = W.Mem(pid)
    obj = mem.q(base + (W.G_CARDSDB - W.IMG_BASE))
    if not obj:
        print("CardsDb singleton is NULL (no FUT session loaded).")
        return 1

    ns = W.nodes(mem, obj) if hasattr(W, "nodes") else P.nodes(mem, obj)
    print("pid=%d  CardsDb=%#x  walked=%d\n" % (pid, obj, len(ns)))

    pairs = collections.Counter()
    states = collections.Counter()
    rows = []
    for n in ns:
        buf = mem.read(n + 0x28, REC_SIZE)
        if not buf or len(buf) < REC_SIZE:
            continue
        resource = P.u32(buf, F_RESOURCE)
        cardtype = P.u8(buf, F_CARDTYPE)
        subtype = P.u8(buf, F_SUBTYPE)
        state = P.u8(buf, F_ITEMSTATE)
        pairs[(subtype, cardtype)] += 1
        states[state] += 1
        rows.append((resource, subtype, cardtype, state))

    print("%-12s %-9s %-9s %s" % ("resource", "subtype", "cardtype", "itemState"))
    for r in sorted(rows):
        print("%-12d %-9d %-9d %d" % r)

    print("\n--- MEASURED cardsubtypeid -> cardtype ---")
    for (sub, ct), n in sorted(pairs.items()):
        want = EXPECTED_CARDTYPE.get(sub)
        if want is None:
            verdict = "no Ghidra prediction for this subtype"
        elif want == ct:
            verdict = "agrees with FUN_1800d8330"
        else:
            verdict = "DISAGREES -- Ghidra said %d" % want
        print("  subtype %-4d -> cardtype %-4d  (%d record(s))  %s" % (sub, ct, n, verdict))

    print("\n--- MEASURED itemState runtime values ---")
    for st, n in sorted(states.items()):
        print("  %-5d  %d record(s)" % (st, n))
    print("\nNOTE: a runtime value only appears here if the client was actually")
    print("served an item in that state. Absence is not evidence of absence.")
    return 0


if __name__ == "__main__":
    sys.exit(main())